Missing Rate Research Articles

A novel MissForest-based missing values imputation approach with recursive feature elimination in medical applications

BackgroundMissing values in datasets present significant challenges for data analysis, particularly in the medical field where data accuracy is crucial for patient diagnosis and treatment. Although MissForest (MF) has demonstrated efficacy in imputation research and recursive feature elimination (RFE) has proven effective in feature selection, the potential for enhancing MF through RFE integration remains unexplored.MethodsThis study introduces a novel imputation method, “recursive feature elimination-MissForest” (RFE-MF), designed to enhance imputation quality by reducing the impact of irrelevant features. A comparative analysis is conducted between RFE-MF and four classical imputation methods: mean/mode, k-nearest neighbors (kNN), multiple imputation by chained equations (MICE), and MF. The comparison is carried out across ten medical datasets containing both numerical and mixed data types. Different missing data rates, ranging from 10 to 50%, are evaluated under the missing completely at random (MCAR) mechanism. The performance of each method is assessed using two evaluation metrics: normalized root mean squared error (NRMSE) and predictive fidelity criterion (PFC). Additionally, paired samples t-tests are employed to analyze the statistical significance of differences among the outcomes.ResultsThe findings indicate that RFE-MF demonstrates superior performance across the majority of datasets when compared to four classical imputation methods (mean/mode, kNN, MICE, and MF). Notably, RFE-MF consistently outperforms the original MF, irrespective of variable type (numerical or categorical). Mean/mode imputation exhibits consistent performance across various scenarios. Conversely, the efficacy of kNN imputation fluctuates in relation to varying missing data rates.ConclusionThis study demonstrates that RFE-MF holds promise as an effective imputation method for medical datasets, providing a novel approach to addressing missing data challenges in medical applications.

A hybrid model for missing traffic flow data imputation based on clustering and attention mechanism optimizing LSTM and AdaBoost

Reliable traffic flow data is not only crucial for traffic management and planning, but also the foundation for many intelligent applications. However, the phenomenon of missing traffic flow data often occurs, so we propose an imputation model for missing traffic flow data to overcome the randomness and instability bands of traffic flow. First, k-means clustering is used to classify road segments with traffic flow belonging to the same pattern into a group to utilize the spatial characteristics of roads fully. Then, the LSTM networks optimized with an attention mechanism are used as the base learner to extract the temporal dependence of the traffic flow. Finally, the AdaBoost algorithm is used to integrate all the LSTM-attention networks into a reinforced learner to impute the missing data. To validate the effectiveness of the proposed model, we use the PeMS dataset for validation, we impute the data with missing data rate from 10 to 60% under three missing modes, and we use multiple baseline models for comparison, which confirms that our proposed model improves the stability and accuracy of imputing the missing data of the traffic flow with different scenarios.

Comparing Right-sided Colon Adenoma and Serrated Polyp Miss Rates with Water Exchange and CO2 Insufflation: A Randomized Controlled Trial

INTRODUCTION: Postcolonoscopy colorectal cancers primarily occur in the right-sided colon because of missed adenomas and serrated polyps (SPs). Water exchange (WE) improves cleanliness and visibility of the right-sided colon. We hypothesized that WE could reduce the right-sided colon adenoma (rAMR) and SP miss rate (rSPMR) compared to standard colonoscopy. METHODS: We randomly assigned 386 colonoscopy patients to insertion with either WE or CO2 insufflation. During the first withdrawal, polypectomies were performed up to the hepatic flexure. A second endoscopist, blinded to the insertion technique, reexamined the right-sided colon. The miss rate was determined by dividing the number of additional adenomas or SPs by the total number detected in both examinations. The primary outcome was the combined rAMR and rSPMR. RESULTS: WE significantly decreased the combined rAMR and rSPMR (22.2% vs 32.2%, P < 0.001) and rSPMR alone (22.5% vs 37.1%, P = 0.002) compared to CO2 insufflation, but not rAMR (21.8% vs 29.8%, P = 0.079). Additionally, WE significantly increased the detection of SP per colonoscopy (SPPC) in the right-sided colon (0.95 ± 1.56 vs 0.50 ± 0.79, P < 0.001). Multivariate logistic regression analysis showed that ≥2 SPs in the right-sided colon was an independent predictor of rSPMR (odds ratio [OR], 3.47; 95% confidence interval [CI], 1.89─6.38), along with a higher right-sided colon Boston Bowel Preparation Scale score (OR, 0.55; 95% CI, 0.32─0.94). CONCLUSIONS: The significant reduction in rSPMR and increase in right-sided colon SPPC suggest that colonoscopy insertion using WE is a valid alternative to CO2 insufflation (Clinical trial registration number: NCT04124393).

OHDLF: A Method for Selecting Orthologous Genes for Phylogenetic Construction and Its Application in the Genus Camellia

Accurate phylogenetic tree construction for species without reference genomes often relies on de novo transcriptome assembly to identify single-copy orthologous genes. However, challenges such as whole-genome duplication (WGD), heterozygosity, gene duplication, and loss can hinder the selection of these genes, leading to limited data for constructing reliable species trees. To address these issues, we developed a new analytical pipeline, OHDLF (Orthologous Haploid Duplication and Loss Filter), which filters orthologous genes from transcript data and adapts parameter settings based on genomic characteristics for further phylogenetic tree construction. In this study, we applied OHDLF to the genus Camellia and evaluated its effectiveness in constructing phylogenetic trees. The results highlighted the pipeline’s ability to handle challenges like high heterozygosity and recent gene duplications by selectively retaining genes with a missing rate and merging duplicates with high similarity. This approach ensured the preservation of informative sites and produced a highly supported consensus tree for Camellia. Additionally, we evaluate the accuracy of the OHDLF phylogenetic trees for different species, demonstrating that the OHDLF pipeline provides a flexible and effective method for selecting orthologous genes and constructing accurate phylogenetic trees, adapting to the genomic characteristics of various plant groups.

The Effect of Regularized Regression and Tree-Based Missing Data Imputation Methods on Classification Performance in High Dimensional Data

Missing data is an important problem in the analysis and classification of high dimensional data. The aim of this study is to compare the effects of four different missing data imputation methods on classification performance in high dimensional data. In this study, missing data imputation methods were evaluated using data sets, whose independent variables between mixed correlated with each other, for binary dependent variable, p=500 independent variables, n=150 units and 1000 times running simulation. Missing data structures were created according to different missing rates. Different datasets were obtained by imputing the missing values using different methods. Regularized regression methods such as lasso and elastic net regression were used for imputation, as well as tree-based methods such as support vector machine and classification and regression trees. At the end of simulation, the classification scores of the methods were obtained by gradient boosting machines and the missing data prediction performances were evaluated according to the distance of these scores from the reference. Our simulation demonstrates that regularized regression methods outperform tree-based methods in classifying high dimensional datasets. Additionally, it was found that the increase in the amount of missing values reduced the classification performance of the methods in high dimensional data.

Prevalence and predictors of near misses and road traffic crashes among long-distance bus drivers in Ghana

IntroductionDespite the critical role of bus drivers in the Ghanaian economy, research on their safety is limited, hindering progress towards reducing road traffic crashes (RTCs) in line with Sustainable Development Goal (SDG) 3.6. This study aimed to investigate the rate of RTCs and near misses among long-distance bus drivers, identifying key predictors.MethodThis cross-sectionl survey sampled 7,315 long-distance bus drivers operating from Accra and Tema to other parts of Ghana and sub-region. Descriptive and binary logistic regression analyses were condcuted using Jamovi software version 2.5.3.ResultsThe study found that 53.7% and 96.9% of bus drivers had at least one RTC and one near-miss two years prior to data collection respectively. Older age, higher education, and extensive driving experience (≥ 21 years) reduced RTC risk, while factors like 6–20 years of experience, individual bus ownership, irregular shifts, long driving hours, sleepiness, job security, low decision authority, and skill discretion increased RTC risk. Near-misses were more likely with public bus ownership, not driving alone, low decision authority, skill discretion, and supervisor support, while high work-family conflict reduced near-miss likelihood.ConclusionConcerted efforts are needed to implement targeted interventions aimed at regulating working hours, enhancing job security and integrating occupationl health and safety standards in the Ghanaian bus transport industry. By prioritizing evidence-based policies and practices, stakeholders can work towards creating safer roads and healthier work environments for bus drivers.

Traffic data imputation via knowledge graph-enhanced generative adversarial network

Traffic data imputation is crucial for the reliability and efficiency of intelligent transportation systems (ITSs), forming the foundation for downstream tasks like traffic prediction and management. However, existing deep learning-based imputation methods struggle with two significant challenges: poor performance under high missing data rates and the limited incorporation of external traffic-related factors. To address these challenges, we propose a novel knowledge graph-enhanced generative adversarial network (KG-GAN) for traffic data imputation. Our approach uniquely integrates external knowledge with traffic spatiotemporal dependencies to improve data imputation quality. Specifically, we construct a fine-grained knowledge graph (KG) that differentiates attributes and relationships of external factors such as points of interest (POI) and weather conditions, facilitating more robust knowledge representation learning. We then introduce a knowledge-aware embedding cell (EM-cell) that merges traffic data with these learned external representations, providing richer inputs for the spatiotemporal GAN. Extensive experiments on a large-scale real-world traffic dataset demonstrate that KG-GAN significantly outperforms state-of-the-art methods under various missing data scenarios. Additionally, ablation studies confirm the superior performance gained from incorporating external knowledge, underscoring the importance of this approach in addressing complex missing data patterns.

The association of pain screening and pain level with suicide among US veterans with comorbid musculoskeletal and bipolar disorder diagnoses

BackgroundHigh suicide rates are documented among persons with bipolar disorder diagnoses and pain diagnoses, but few studies have examined the association of pain with suicide mortality in individuals with comorbid pain and bipolar disorders. This study assessed the association of pain screening and pain severity with suicide mortality among veterans with comorbid bipolar and musculoskeletal disorder (MSD) diagnoses. MethodsA retrospective cohort study was conducted on 168,021 patients within the Veterans Health Administration (VHA) who received an MSD diagnosis from 2000 to 2015 and had a bipolar disorder diagnosis. Pain severity, comorbidities, demographics, and suicide mortality were extracted from VHA databases. Poisson regression examined relative risk of suicide by the presence pain screening and pain severity ratings. ResultsPain was assessed in 72.73 % of veterans. Suicide risk was greater in those not assessed (0.98 % versus 0.77 % in assessed group). However, this result did not persist after adjusting for covariates (RR = 1.06). Among those assessed, higher suicide risk was associated with moderate (RR = 1.10), severe pain (RR = 1.06), and no pain (reference) relative to mild pain (RR = 0.99). Major depression, substance use disorders, and prescribed opioids and benzodiazepines increased risk. LimitationsData were obtained from medical records; diagnoses were not confirmed via formal assessment, and no information was available on actual medication use or purpose. Over 25 % of the sample were missing pain severity ratings, which could have affected results. ConclusionsSuicide risk factors among persons with bipolar disorder are complex and multifactorial. Providers should prioritize suicide prevention efforts following new onset or worsening pain.

A-196 Using Aggregated Proficiency Testing Results to Identify Systematic Error

Abstract Background Proficiency testing (PT) should identify systematic errors, which are likely to recur. The Clinical Laboratory Improvement Amendments of 1988 (CLIA) acceptance limits (ALs) include 3SD and concentration limits. We investigated the ability of PT to detect systematic error, especially as affected by the AL type. Methods Ungradable, duplicate, and irregular scores were removed from 2008 to 2018 CLIA laboratory PT results. For analytes that had at least 200 events and at least 200 participating laboratories, we calculated the overall sample miss rate, the unsatisfactory event rate, i.e., less than 80, and event rates for Score 100 to 0. We used paired t-tests and the Wilcoxon Signed-Rank test to compare miss rates and unsatisfactory event rates between short- and long-term participants. We used the binomial distribution to model the expected event scores under the assumption that all errors may be independent (random); see Figure. For each analyte and each of the possible PT scores, we compared observed event scores with their expected values as a ratio. Results 40,596 laboratories produced 15,140,128 event scores for 75 analytes. The distribution of event scores was over-weighted with higher-order event errors (Score 0 - Score 60) compared to the predicted distribution. Miss rates and unsatisfactory rates were significantly higher for short-term laboratories. Plotting the log ratio of observed versus expected rates for event scores showed that the degree of systematic effect increased progressively for each analyte. The magnitude of the effect was less for 3SD ALs. Conclusions In an event, PT errors are often dependent. Concentration and 3SD ALs were similarly sensitive to detect systematic error; both were more sensitive than ALs for qualitative or dilution assays. Expressing systematic error using PT data could help to identify remediable analytical issues.

A shaping two-stage anomaly data recovery method based on multi-norm joint optimization under energy internet

With the deep integration of energy networks and information communication technology, the involvement of hardware and environmental factors intensifies the challenges to data security. Prior research focuses predominantly on interpolating random missing data under the assumption of no data damage, without providing strategies for addressing continuous data loss. To overcome, we proposes a novel two-stage data recovery method, that includes three-fold ideas: (1) it introduces the matrix shaping differentiation strategy to realize the differentiated recovery of random and continuous missing data; (2) it designs a two-stage data recovery approach, in which the first stage constructs a random missing data recovery objective function by eliminating the consecutive missing data columns, and the second stage performs temporal pre-interpolation backfilling on the eliminated columns and introduces the temporal correlation to construct the temporal smoothing objective function; and (3) it develops an alternating direction method of multipliers-based iterative solution algorithm. Experimental results on four public datasets demonstrate that our model is superior to five state-of-the-art and classical methods in terms of random missing error rate, random column missing error rate, root mean square error, and mean absolute error.

“Anything that would help is a positive development”: feasibility, tolerability, and user experience of smartphone-based digital phenotyping for people with and without type 2 diabetes

BackgroundDigital phenotyping, the in-situ collection of passive (phone sensor) and active (daily surveys) data using a digital device, may provide new insights into the complex relationship between daily behaviour and mood for people with type 2 diabetes. However, there are critical knowledge gaps regarding its use in people with type 2 diabetes. This study assessed feasibility, tolerability, and user experience of digital phenotyping in people with and without type 2 diabetes after participation in a 2-month digital phenotyping study in Ireland. At study completion, participants rated methodology elements from “not a problem” to a “serious problem” on a 5-point scale and reported their comfort with the potential future use of digital phenotyping in healthcare, with space for qualitative expansion.ResultsEighty-two participants completed baseline. Attrition was 18.8%. Missing data ranged from 9–44% depending on data stream. Sixty-eight participants (82.9%) completed the user experience questionnaire (51.5% with type 2 diabetes; 61.8% female; median age-group 50–59). Tolerability of digital phenotyping was high, with “not a problem” being selected 76.5%—89.7% of the time across questions. People with type 2 diabetes (93.9%) were significantly more likely to be comfortable with their future healthcare provider having access to their digital phenotyping data than those without (53.1%), χ2 (1) = 14.01, p = < .001. Free text responses reflected a range of positive and negative experiences with the study methodology.ConclusionsAn uncompensated, 2-month digital phenotyping study was feasible among people with and without diabetes, with low attrition and reasonable missing data rates. Participants found digital phenotyping to be acceptable, and even enjoyable. The potential benefits of digital phenotyping for healthcare may be more apparent to people with type 2 diabetes than the general population.

A high-efficiency local and global detector for diatom-based drowning diagnosis

Detection and classification of diatoms has been considered as the most effective method for the drowning diagnosis in forensic practice. However, challenges (such as limited samples, complex background interference, and detecting multiple objects) remain during detection and classification. To address these challenges, we propose a MDMS (multi-scale dynamic multi-head self-attention)-based deep learning network for the diatom-based drowning diagnosis. This framework uses TFFT (two-stage full fine-tuning training strategy) to solve the challenges of limited diatom samples and detecting multiple objects, and builds a multi-scale dynamic multi-head self-attention module to extract the local and global features of diatom images. In the meantime, we introduce an online hard example mining strategy to attenuate the complex background interference. The experimental results show that the proposed framework can effectively reduce the missing and false detection rates of diatom objects, with the mAP (mean Average Precision) reaching 92.434%, which is better than the result using the mainstream methods.

A lightweight CNN-Transformer network for pixel-based crop mapping using time-series Sentinel-2 imagery

Deep learning approaches have provided state-of-the-art performance in crop mapping. Recently, several studies have combined the strengths of two dominant deep learning architectures, Convolutional Neural Networks (CNNs) and Transformers, to classify crops using remote sensing images. Despite their success, many of these models utilize patch-based methods that require extensive data labeling, as each sample contains multiple pixels with corresponding labels. This leads to higher costs in data preparation and processing. Moreover, previous methods rarely considered the impact of missing values caused by clouds and no-observations in remote sensing data. Therefore, this study proposes a lightweight multi-stage CNN-Transformer network (MCTNet) for pixel-based crop mapping using time-series Sentinel-2 imagery. MCTNet consists of several successive modules, each containing a CNN sub-module and a Transformer sub-module to extract important features from the images, respectively. An attention-based learnable positional encoding (ALPE) module is designed in the Transformer sub-module to capture the complex temporal relations in the time-series data with different missing rates. Arkansas and California in the U.S. are selected to evaluate the model. Experimental results show that the MCTNet has a lightweight advantage with the fewest parameters and memory usage while achieving the superior performance compared to eight advanced models. Specifically, MCTNet obtained an overall accuracy (OA) of 0.968, a kappa coefficient (Kappa) of 0.951, and a macro-averaged F1 score (F1) of 0.933 in Arkansas, and an OA of 0.852, a Kappa of 0.806, and an F1 score of 0.829 in California. The results highlight the importance of each component of the model, particularly the ALPE module, which enhanced the Kappa of MCTNet by 4.2% in Arkansas and improved the model’s robustness to missing values in remote sensing data. Additionally, visualization results demonstrated that the features extracted from CNN and Transformer sub-modules are complementary, explaining the effectiveness of the MCTNet.

Dynamic attention aggregated missing spatial–temporal data imputation for traffic speed prediction

Traffic speed forecasting is indispensable in Intelligent Traffic Systems (ITS). The missing traffic speed due to sensors’ failure may hamper the accurate prediction of traffic speed. Different kinds of imputation techniques were employed to replace the missing values considering the spatial and temporal information. However, previous techniques did not consider the dynamic contribution of the neighboring nodes and failed to capture the complex properties of traffic speed data precisely. The architecture of the previous imputation models was infeasible in real-time applications in terms of scalability. Therefore, a novel machine learning-based imputation technique is proposed to generate the missing traffic speed using the spatial–temporal information and dynamic contribution of neighboring nodes. A new dataset considering the characteristics of the traffic speed data is generated from the existing dataset by exploiting the non-missing traffic speed of neighboring nodes. The newly formed dataset helps in the estimation of the dynamic contribution of the neighboring nodes using the linear regression algorithm. The non-missing speeds and estimated dynamic contribution of neighboring nodes assist in calculating the missing speeds of a node. Furthermore, a novel deep learning-based prediction model is introduced to forecast traffic speed accurately. The prediction model contains graph structure learning, a deep graph neural network with skip connections, GRU, multi-head soft attention, and a fully connected neural network. Graph structure learning is proposed based on the distribution of the congestion to represent the graph in an efficient way. The deep graph neural network with skip connections and GRU help in capturing topological and temporal information, respectively. The multi-head soft attention is designed to focus on every relevant temporal information at each time step for capturing global traffic information. Finally, the extracted spatial–temporal features are fed into the fully connected neural network to predict the traffic speed. The proposed imputation and prediction models are evaluated on two real-time datasets, METR-LA and PeMSD7 datasets, under different missing rates of different missing patterns. The proposed framework generates possible spatial–temporal information for efficient traffic congestion management and evicts erroneous issues in real-time.

An MRI multi-sequence feature imputation and fusion mutual-aid model based on sequence deletion for differentiation of high-grade from low-grade glioma

To evaluate the performance of magnetic resonance imaging (MRI) multi-sequence feature imputation and fusion mutual model based on sequence deletion in differentiating high-grade glioma (HGG) from low-grade glioma (LGG). We retrospectively collected multi-sequence MR images from 305 glioma patients, including 189 HGG patients and 116 LGG patients. The region of interest (ROI) of T1-weighted images (T1WI), T2-weighted images (T2WI), T2 fluid attenuated inversion recovery (T2_FLAIR) and post-contrast enhancement T1WI (CE_T1WI) were delineated to extract the radiomics features. A mutual-aid model of MRI multi-sequence feature imputation and fusion based on sequence deletion was used for imputation and fusion of the feature matrix with missing data. The discriminative ability of the model was evaluated using 5-fold cross-validation method and by assessing the accuracy, balanced accuracy, area under the ROC curve (AUC), specificity, and sensitivity. The proposed model was quantitatively compared with other non-holonomic multimodal classification models for discriminating HGG and LGG. Class separability experiments were performed on the latent features learned by the proposed feature imputation and fusion methods to observe the classification effect of the samples in twodimensional plane. Convergence experiments were used to verify the feasibility of the model. For differentiation of HGG from LGG with a missing rate of 10%, the proposed model achieved accuracy, balanced accuracy, AUC, specificity, and sensitivity of 0.777, 0.768, 0.826, 0.754 and 0.780, respectively. The fused latent features showed excellent performance in the class separability experiment, and the algorithm could be iterated to convergence with superior classification performance over other methods at the missing rates of 30% and 50%. The proposed model has excellent performance in classification task of HGG and LGG and outperforms other non-holonomic multimodal classification models, demonstrating its potential for efficient processing of non-holonomic multimodal data.

What's the harm? Results of an active surveillance adverse event reporting system for chiropractors and physiotherapists.

This prospective, community-based, active surveillance study aimed to report the incidence of moderate, severe, and serious adverse events (AEs) after chiropractic (n = 100) / physiotherapist (n = 50) visit in offices throughout North America between October-2015 and December-2017. Three content-validated questionnaires were used to collect AE information: two completed by the patient (pre-treatment [T0] and 2-7 days post-treatment [T2]) and one completed by the provider immediately post-treatment [T1]. Any new or worsened symptom was considered an AE and further classified as mild, moderate, severe or serious. From the 42 participating providers (31 chiropractors; 11 physiotherapists), 3819 patient visits had complete T0 and T1 assessments. The patients were on average 50±18 years of age and 62.5% females. Neck/back pain was the most common presenting condition (70.0%) with 24.3% of patients reporting no condition/preventative care. From the patients visits with a complete T2 assessment (n = 2136 patient visits, 55.9%), 21.3% reported an AE, of which: 7.9% were mild, 6.2% moderate, 3.7% severe, 1.5% serious, and 2.0% had missing severity rating. The most common symptoms reported with moderate or higher severity were discomfort/pain, stiffness, difficulty walking and headache. This study provides valuable information for patients and providers regarding incidence and severity of AEs following patient visits in multiple community-based professions. These findings can be used to inform patients of what AEs may occur and future research opportunities can focus on mitigating common AEs.

Natural generative noise diffusion model imputation

Imputation is a critical method for enhancing dataset quality, essential for ensuring accurate analysis and insights. This research proposes an advanced imputation algorithm utilizing a Diffusion Model enhanced with Perlin noise generation. We introduce Perlin noise at each step of the diffusion process and incorporate a cosine scheduler to optimize performance. Our approach demonstrates improvements in imputing non-normal data, validated through tests on ten real datasets. Due to the substantial slope of distribution properties produced by perlin noise, noisy data gradually contaminates nonnormally distributed data, making it more like to a Gaussian distribution. The Perlin noise distribution increases the normality of the noisy data provided noise when it enters the deep neural process of diffusion imputation. We assess our proposed approach by simulating the missing data rate using three scenarios: Missing Completely at Random (MCAR), Missing Not At Random (MNAR), and Missing at Random (MAR). Every case is handled similarly, with 20 % to 80 % missing data. Compared to other deep learning imputation methods, our proposed methods and improvements contribute to lowering the RMSE value up to 10 % on non-normal distributed data imputation.

Exploring the impact of temporally contiguous action effect on action control performance in typical development and attention-deficit/hyperactivity disorder.

Previous work shows a reinforcing impact of action effect on behavior, independent of other reinforces such as positive outcomes or task success. Action-effect temporal contiguity plays an important role in such a reinforcing effect, possibly indicating a motor-based evaluation of their causal relationship. In the present study, we aimed to negate the reinforcing impact of an immediate action effect with task success by designing a task where red and green circle stimuli rapidly descended on the screen. Participants were instructed to respond only when a specific sequence of colored stimuli matched a predefined response rule. The temporal contiguity between the response and a perceptual effect was manipulated. We initially hypothesized an increased action tendency resulting in higher false alarm rates in the immediate (compared to 400 ms lag) action-effect condition. We also expected this pattern to be more pronounced in attention-deficit/hyperactivity disorder compared to typically developing individuals. Contrary to our expectations, results from three experiments showed a consistent pattern of a lower false alarm rate in the immediate compared to the 400 ms lag effect condition across both attention-deficit/hyperactivity disorder and typically developing groups. Additionally, while action-effect temporal contiguity did not significantly alter the overall rate of misses, we observed earlier improvements in both misses and false alarms in the immediate condition during the first blocks. Possible explanations for the complex impact of action effect on action tendency and action control are discussed. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Using AI to Identify Unremarkable Chest Radiographs for Automatic Reporting.

Background Radiology practices have a high volume of unremarkable chest radiographs and artificial intelligence (AI) could possibly improve workflow by providing an automatic report. Purpose To estimate the proportion of unremarkable chest radiographs, where AI can correctly exclude pathology (ie, specificity) without increasing diagnostic errors. Materials and Methods In this retrospective study, consecutive chest radiographs in unique adult patients (≥18 years of age) were obtained January 1-12, 2020, at four Danish hospitals. Exclusion criteria included insufficient radiology reports or AI output error. Two thoracic radiologists, who were blinded to AI output, labeled chest radiographs as "remarkable" or "unremarkable" based on predefined unremarkable findings (reference standard). Radiology reports were classified similarly. A commercial AI tool was adapted to output a chest radiograph "remarkableness" probability, which was used to calculate specificity at different AI sensitivities. Chest radiographs with missed findings by AI and/or the radiology report were graded by one thoracic radiologist as critical, clinically significant, or clinically insignificant. Paired proportions were compared using the McNemar test. Results A total of 1961 patients were included (median age, 72 years [IQR, 58-81 years]; 993 female), with one chest radiograph per patient. The reference standard labeled 1231 of 1961 chest radiographs (62.8%) as remarkable and 730 of 1961 (37.2%) as unremarkable. At 99.9%, 99.0%, and 98.0% sensitivity, the AI had a specificity of 24.5% (179 of 730 radiographs [95% CI: 21, 28]), 47.1% (344 of 730 radiographs [95% CI: 43, 51]), and 52.7% (385 of 730 radiographs [95% CI: 49, 56]), respectively. With the AI fixed to have a similar sensitivity as radiology reports (87.2%), the missed findings of AI and reports had 2.2% (27 of 1231 radiographs) and 1.1% (14 of 1231 radiographs) classified as critical (P = .01), 4.1% (51 of 1231 radiographs) and 3.6% (44 of 1231 radiographs) classified as clinically significant (P = .46), and 6.5% (80 of 1231) and 8.1% (100 of 1231) classified as clinically insignificant (P = .11), respectively. At sensitivities greater than or equal to 95.4%, the AI tool exhibited less than or equal to 1.1% critical misses. Conclusion A commercial AI tool used off-label could correctly exclude pathology in 24.5%-52.7% of all unremarkable chest radiographs at greater than or equal to 98% sensitivity. The AI had equal or lower rates of critical misses than radiology reports at sensitivities greater than or equal to 95.4%. These results should be confirmed in a prospective study. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Yoon and Hwang in this issue.

The relationship between trust in federal oversight of vaccine safety and willingness to participate in COVID-19 clinical trials: a repeated measures study of Philadelphia residents (September 2021 – March 2023)

BackgroundThe Coronavirus Disease 2019 (COVID-19) pandemic precipitated an urgent need for clinical trials to discover safe and efficacious treatments. We examined how COVID-19 experiences, clinical trial awareness, and trust in the vaccine safety process were associated with willingness to participate in COVID-19 clinical trials. The objective was to investigate the relationship between trust in federal oversight of vaccine safety and willingness to participate in clinical trials for COVID-19 treatment across four distinct time points over an 18-month period during the COVID-19 pandemic.MethodsWe used four waves of data collected from September 2021 to March 2023 among 582 Philadelphia residents (with a missing data rate of 0.9%). Generalized estimating equations estimated the association between willingness to participate in COVID-19 clinical trials and participants’ trust in the federal government’s oversight of COVID-19 vaccine safety, COVID-19-related variables (COVID-19 related health challenges, history of COVID-19 infection), awareness of clinical trials and how to enroll in them, and sociodemographic characteristics (age, race/ethnicity, sexual orientation, gender, parental status, education, and insurance).ResultsOn average, willingness to participate in a COVID-19 clinical trial was positively associated with greater trust in the federal government’s oversight of vaccine safety [β = 0.34, 95% confidence interval (CI): 0.15–0.53], having COVID-19 (β = 0.40, 95% CI: 0.08–0.73), awareness of clinical trials (β = 0.38, 95% CI: 0.04–0.73), and knowledge of how to enroll (β = 0.83, 95% CI: 0.44–1.23). Among sociodemographic characteristics, race/ethnicity (p = 0.001) and gender (p = 0.018) were identified as predictors for COVID-19 trial willingness.ConclusionWillingness to participate in clinical trials may be bolstered by strengthening the public’s trust in the federal government’s role within vaccine safety oversight, increasing the perceived relevance of clinical trials to individuals’ health and well-being, and offering tailored information to educate diverse communities about ongoing trials and how to enroll in them.